Data

Load the Data and choose the variables that I want to take a look at.

# Read the data set
climate <- read_excel("climate_change_download_0.xls",sheet = 1)
countries <- read_excel("climate_change_download_0.xls",sheet = 2)

# choose variables that I want to take a look at 
chosen <- climate %>% filter(`Series code` %in%  c("NY.GDP.MKTP.CD","NY.GNP.PCAP.CD","EG.USE.PCAP.KG.OE","SP.POP.TOTL","EN.ATM.CO2E.PC","EN.ATM.CO2E.KT","SP.POP.GROW"))
chosen <- chosen[,-c(5,6,27,28)]

head(chosen)

Clean the data

chosen_melt <- melt(chosen,id=c("Country code","Country name","Series code","Series name"))
colnames(chosen_melt) <- c("CountryCode","CountryName","SeriesCode","SeriesName","Year","Value")

# replace all ".." cells with NA
chosen_melt <- chosen_melt %>% replace_with_na(replace = list(Value = ".."))
chosen_melt$Value <- as.numeric(chosen_melt$Value)%>%round(2)

# Make each variable as one column
chosen_spread <- spread(chosen_melt,key=SeriesName,value = Value)
chosen_spread <- chosen_spread[,-c(1,3)]

head(chosen_melt)

Take a look at each variable one by one

Energy use per capita (kilograms of oil equivalent)

energy <- chosen_melt%>%select(CountryName,SeriesName,Year,Value)%>%filter(SeriesName == "Energy use per capita (kilograms of oil equivalent)")

# Take a look at the ggplot
ggplot(energy,aes(Year,Value,color = Year)) + geom_point() + ylab("Energy Use")
## Warning: Removed 1526 rows containing missing values (geom_point).

# Total World Energy use per capita (kilograms of oil equivalent)
energy_tot <- aggregate(energy[,4],by = list(energy$Year),FUN = sum,na.rm=TRUE)
colnames(energy_tot) <- c("Year","EnergyUse")

# Take a look at how energy usage changed over decades
energy_tot$Year <- as.numeric(as.character(energy_tot$Year))
plotly_build(ggplot(energy_tot) + geom_line(aes(Year,EnergyUse),stat = "identity") + geom_point(aes(Year,EnergyUse)) + xlab("Year") + ylab("Energy Use"))

CO2 emissions, total (KtCO2)

CO2Total <- chosen_melt%>%select(CountryName,SeriesName,Year,Value)%>%filter(SeriesName == "CO2 emissions, total (KtCO2)",Year != 2009)

# Take a look at the ggplot
ggplot(CO2Total,aes(Year,Value,color = Year)) + geom_point() + ylab("World CO2 Emissions(kt)")
## Warning: Removed 444 rows containing missing values (geom_point).

# World CO2 emissions, total (KtCO2)
CO2_tot <- aggregate(CO2Total[,4],by = list(CO2Total$Year),FUN = sum,na.rm=TRUE)
colnames(CO2_tot) <- c("Year","World CO2 Emissions(kt)")

# Take a look at how total CO2 emissions changed over decades
CO2_tot$Year <- as.numeric(as.character(CO2_tot$Year))
plotly_build(ggplot(CO2_tot) + geom_line(aes(Year,`World CO2 Emissions(kt)`),stat = "identity") + geom_point(aes(Year,`World CO2 Emissions(kt)`)) + xlab("Year") + ylab("World CO2 Emissions(kt)"))

CO2 emissions per capita (metric tons)

CO2_capita <- chosen_melt%>%select(CountryName,SeriesName,Year,Value)%>%filter(SeriesName == "CO2 emissions per capita (metric tons)",Year != 2009)

# Take a look at the ggplot
ggplot(CO2_capita,aes(Year,Value,color = Year)) + geom_point() + ylab("World CO2 Emissions per capita")
## Warning: Removed 447 rows containing missing values (geom_point).

# World CO2 emissions per capita (metric tons)
CO2_capita_tot <- aggregate(CO2_capita[,4],by = list(CO2_capita$Year),FUN = sum,na.rm=TRUE)
colnames(CO2_capita_tot) <- c("Year","World CO2 Emissions per capita")

# Take a look at how world CO2 emissions per capita changed over decades
CO2_capita_tot$Year <- as.numeric(as.character(CO2_capita_tot$Year))
plotly_build(ggplot(CO2_capita_tot) + geom_line(aes(Year,`World CO2 Emissions per capita`),stat = "identity") + geom_point(aes(Year,`World CO2 Emissions per capita`)) + xlab("Year") + ylab("World CO2 Emissions per capita"))

GDP ($)

GDP <- chosen_melt%>%select(CountryName,SeriesName,Year,Value)%>%filter(SeriesName == "GDP ($)")

# Take a look at the ggplot
ggplot(GDP,aes(Year,Value,color = Year)) + geom_point()
## Warning: Removed 495 rows containing missing values (geom_point).

GNI per capita (Atlas $)

GNI <- chosen_melt%>%select(CountryName,SeriesName,Year,Value)%>%filter(SeriesName == "GNI per capita (Atlas $)")

# Take a look at the ggplot
ggplot(GNI,aes(Year,Value,color = Year)) + geom_point()
## Warning: Removed 726 rows containing missing values (geom_point).

Population growth (annual %)

popgrowth <- chosen_melt%>%select(CountryName,SeriesName,Year,Value)%>%filter(SeriesName == "Population growth (annual %)")

# Take a look at the ggplot
ggplot(popgrowth,aes(Year,Value,color = Year)) + geom_point() + ylab("Population Growth")
## Warning: Removed 23 rows containing missing values (geom_point).

# World Population
popgrowth_tot <- aggregate(popgrowth[,4],by = list(popgrowth$Year),FUN = mean,na.rm=TRUE)
colnames(popgrowth_tot) <- c("Year","PopulationGrowth")

# Take a look at how world population changed over decades
popgrowth_tot$Year <- as.numeric(as.character(popgrowth_tot$Year))
plotly_build(ggplot(popgrowth_tot) + geom_line(aes(Year,`PopulationGrowth`),stat = "identity") + geom_point(aes(Year,`PopulationGrowth`))+ xlab("Year") + ylab("Population Growth"))

Population

pop <- chosen_melt%>%select(CountryName,SeriesName,Year,Value)%>%filter(SeriesName == "Population")

# Take a look at the ggplot
ggplot(pop,aes(Year,Value,color = Year)) + geom_point() + ylab("Population")
## Warning: Removed 19 rows containing missing values (geom_point).

# World Population
pop_tot <- aggregate(pop[,4],by = list(pop$Year),FUN = sum,na.rm=TRUE)
colnames(pop_tot) <- c("Year","Population")

# Take a look at how world population changed over decades
pop_tot$Year <- as.numeric(as.character(pop_tot$Year))
plotly_build(ggplot(pop_tot) + geom_line(aes(Year,`Population`),stat = "identity") + geom_point(aes(Year,`Population`)) + xlab("Year") + ylab("Population"))

It is interesting finding that though the world population is growing every year, the annual population growth rate has a decreasing trend.

Take a look at population and CO2 emissions

pop_co2 <- chosen_melt%>%filter(SeriesName %in% c("CO2 emissions, total (KtCO2)","Population"),Year != 2009)%>%spread(key=SeriesName,value = Value)

pop_world <- aggregate(pop_co2[,5:6],by = list(pop_co2$Year,pop_co2$SeriesCode),FUN = sum,na.rm=TRUE)
colnames(pop_world) <- c("Year","SeriesCode","CO2","Population")

pop_world$`CO2` <- as.numeric(as.character(pop_world$CO2))
pop_world$Population <- as.numeric(as.character(pop_world$Population))

ggplot(data = pop_world) + stat_summary(mapping = aes(y = Population, x = Year,color = Year),fun.ymin = min,fun.ymax = max,fun.y = median)

ggplot(data = pop_world) + stat_summary(mapping = aes(y = CO2, x = Year,color = Year),fun.ymin = min,fun.ymax = max,fun.y = median)